Growing evidence suggests that certain racial/ethnic minority groups experience the highest incidence of Alzheimer?s disease and related dementias (ADRD) risk in the US. Longitudinal studies with diverse samples that collect social and behavioral measures in early and midlife hold exceptional promise for identifying modifiable protective factors for cognitive health and life course pathways of ADRD risk. Social relations, a multidimensional concept with high intervention potential, have been found to predict cognitive health and ADRD, though these associations may differ across race/ethnicity. Still not identified are: (1) how early and mid-life social relations are associated with ADRD risk in different racial/ethnic groups; and (2) specific aspects of social relations that modify ADRD risk. Characterizing how social relations vary as a function of race/ethnicity and clarifying unique links to ADRD risk will advance the field in terms of theory and intervention. This study will leverage three waves of the Social Relations Study plus a fourth wave of the >65 sample (currently in the field: N=330) to examine social relations and ADRD risk across racial/ethnic groups. We will collect a fourth wave of panel data aged <65 (N=244) and a new, representative Detroit metropolitan area sample (N=3400), using the original design, adding Arab Americans to parallel data collection now in the field. Specific aims are: Aim 1: Examine secular trends in social relations and health by race/ethnicity in two adult lifespan regionally representative cohorts. Aim 2: Identify aspects of social relations that have the greatest effects on ADRD risk among blacks, whites and Arab Americans in mid- and late- life. Aim 3: Identify longitudinal associations between social relations across the life course and ADRD risk among blacks and whites. We capitalize on an existing longitudinal cohort study of detailed social relations over ~30 years in a diverse lifespan sample (i.e., from age 8 to 93 in Wave 1). Adding cognitive and genetic measures, as well as extending racial/ethnic group comparisons to include Arab Americans in a new regionally representative sample (age 35+) provides a novel opportunity to study modifiable factors in midlife for ADRD risk. Findings will provide key information to develop strategies using the influential resource of social relations to reduce disparities. Further, the proposed study sets the stage for a newly representative longitudinal study of racial/ethnic disparities in ADRD risk.